Merge lp:~vila/otto/lxc-alpha2 into lp:otto
Proposed by
Vincent Ladeuil
Status: | Merged |
---|---|
Approved by: | Didier Roche-Tolomelli |
Approved revision: | 257 |
Merged at revision: | 257 |
Proposed branch: | lp:~vila/otto/lxc-alpha2 |
Merge into: | lp:otto |
Diff against target: |
12 lines (+4/-1) 1 file modified
lxc.defaults/fstab (+4/-1) |
To merge this branch: | bzr merge lp:~vila/otto/lxc-alpha2 |
Related bugs: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Didier Roche-Tolomelli | Approve | ||
Review via email: mp+192347@code.launchpad.net |
Description of the change
After stgraber testing with lxc-alpha1 and lxc-alpha2 on qa-intel-4000:
<stgraber> oh, I think I begin to understand what's going on... so because of what's pretty likely to be an LXC bug, the otto containers don't switch profile, they stay on the host one and so when apparmor starts in the container, it overwrites some profiles
<stgraber> which explains why LXC works as expected before otto starts and the world is broken afterwards
To post a comment you must log in.
tested on qa-intel-4000 by building a new container and using it in http:// 10.97.0. 1:8080/ job/autopilot- trusty- daily_release/ label=qa- intel-4000/ 32/console
which sil2100 said it confirms the fix